Performance Evaluation and Energy Consumption for DVFS Processor

被引:0
|
作者
El Mahjoub, Youssef Ait [1 ,2 ]
Fourneau, Jean-Michel [1 ]
Castel-Taleb, Hind [2 ]
机构
[1] Univ Paris Saclay, DAVID UVSQ, Paris, France
[2] Inst Polytech Paris, Telecom SudParis, SAMOVAR, Paris, France
关键词
DVFS; Performance evaluation; Energy consumption; Stochastic comparison;
D O I
10.1007/978-3-030-91825-5_10
中图分类号
TP301 [理论、方法];
学科分类号
081202 ;
摘要
Dynamic Voltage and Frequency Scaling (DVFS) involves adjusting the speed to match power consumption and performance requirements of the system. This technology has been shown to provide major power and energy savings in many system components (processor cores, memory system ...). In this work, we study the DVFS mechanism for a processor that implements various levels of speed and power (we call these levels "Pstates"). We model the system as a birth-death process and we compare different configurations, using stochastic comparison, in order to evaluate the impact on the response time and power (and energy) consumption. In the case of two Pstates, we proposed a closed form for the steady state distribution and we derived a cost function that uses both performance and energy consumption. Finally, we derive an algorithm that suggests a threshold which minimizes the cost function.
引用
收藏
页码:165 / 180
页数:16
相关论文
共 50 条
  • [21] Some observations on optimal frequency selection in DVFS-based energy consumption minimization
    Rizvandi, Nikzad Babaii
    Taheri, Javid
    Zomaya, Albert Y.
    JOURNAL OF PARALLEL AND DISTRIBUTED COMPUTING, 2011, 71 (08) : 1154 - 1164
  • [22] Modeling Energy Consumption of Virtual Machines in DVFS-Enabled Cloud Data Centers
    Mao, Jianzhou
    Bhattacharya, Tathagata
    Peng, Xiaopu
    Cao, Ting
    Qin, Xiao
    2020 IEEE 39TH INTERNATIONAL PERFORMANCE COMPUTING AND COMMUNICATIONS CONFERENCE (IPCCC), 2020,
  • [23] PAMPAR: A new parallel benchmark for performance and energy consumption evaluation
    Garcia, Adriano Marques
    Schepke, Claudio
    Girardi, Alessandro
    CONCURRENCY AND COMPUTATION-PRACTICE & EXPERIENCE, 2020, 32 (20):
  • [24] Performance Evaluation of Techniques for Identifying Abnormal Energy Consumption in Buildings
    Gaur, Megha
    Makonin, Stephen
    Bajic, Ivan, V
    Majumdar, Angshul
    IEEE ACCESS, 2019, 7 : 62721 - 62733
  • [25] THE EVALUATION OF THE PERFORMANCE OF A GLASS MELTING FURNACE IN TERMS OF ENERGY CONSUMPTION
    Tapasa, Kanit
    Meechoowas, Ekarat
    Naknikham, Usuma
    Jitwatcharakomol, Tepiwan
    SURANAREE JOURNAL OF SCIENCE AND TECHNOLOGY, 2014, 21 (01): : 15 - 20
  • [26] Investigation and Evaluation of Energy Consumption Performance for Hospital Buildings in China
    Ji, Ru
    Qu, Shilin
    SUSTAINABILITY, 2019, 11 (06)
  • [27] Evaluation of performance and energy consumption of an impinging stream dryer for paddy
    Nimmol, Chatchai
    Devahastin, Sakamon
    APPLIED THERMAL ENGINEERING, 2010, 30 (14-15) : 2204 - 2212
  • [28] Evaluation of the Energy Consumption Model Performance for Electric Vehicles in SUMO
    Sagaama, Insaf
    Kchiche, Amine
    Trojet, Wassim
    Kamoun, Farouk
    2019 IEEE/ACM 23RD INTERNATIONAL SYMPOSIUM ON DISTRIBUTED SIMULATION AND REAL TIME APPLICATIONS (DS-RT), 2019, : 59 - 66
  • [29] Performance Evaluation of Energy Consumption for Different DTN Routing Protocols
    Spaho, Evjola
    Dhoska, Klodian
    Bylykbashi, Kevin
    Barolli, Leonard
    Kolici, Vladi
    Takizawa, Makoto
    ADVANCES IN NETWORK-BASED INFORMATION SYSTEMS, NBIS-2018, 2019, 22 : 122 - 131
  • [30] The Impact of GPU DVFS on the Energy and Performance of Deep Learning: an Empirical Study
    Tang, Zhenheng
    Wang, Yuxin
    Wang, Qiang
    Chu, Xiaowen
    E-ENERGY'19: PROCEEDINGS OF THE 10TH ACM INTERNATIONAL CONFERENCE ON FUTURE ENERGY SYSTEMS, 2019, : 315 - 325